This week in BAN7002 we explored using linear regressions to perform estimations on datasets. We learned two new commands: PROC REG and T-TEST.
I found this to be very helpful in testing hypotheses when dealing with data and using the results to compare against a baseline. We used this concept to perform comparative regressions across multiple datasets that contained information about red and white wines, specifically using a TEST data set to compare its accuracy against a SCORE (baseline) data set.
This concept will be extraordinarily helpful in the data analysis world in the future when performing benchmark analyses. This provides much more insight than the typical 5 number summer (mean, median, mode, max, min) that only provides very high level comparison metrics in analyzing a dataset against a baseline.
